Autonomous Driving Reference Line Stitching via Curvature Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing smoothing algorithms for generating reference lines in autonomous driving vehicles often fail to produce smooth connections between segments, especially when there are large gaps in curvature, leading to discontinuous stitch points, particularly at high speeds.
Innovation Solution
A computer-implemented method determines the smallest curvature points within defined distances along path segments to generate smooth reference lines by connecting them using quadratic programming optimization, ensuring continuity and smoothness of the stitched reference lines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional smoothing algorithms (e.g., Quadratic Programming) are used to generate reference lines, then the reference line can be generated, but the stitch point between two reference lines becomes discontinuous when there is a large gap between adjacent curvatures
Solution Approach 1:
The patent applies preliminary action by pre-selecting reference points with smallest curvature values before generating reference lines. By identifying and marking these optimal points in advance along the path segments, the system ensures that when reference lines are stitched together, they connect at points that minimize curvature discontinuities, thereby maintaining smoothness and continuity across segment boundaries
Solution Approach 2:
The patent changes the selection criterion for reference points from arbitrary or uniform spacing to selecting points with smallest curvature values. This parameter change in the selection strategy ensures that reference lines are generated and connected at optimal locations that minimize curvature gaps, resolving the discontinuity issue at stitch points while maintaining overall smoothness
2Productivity
If reference lines are generated and stitched together for high-speed autonomous driving, then navigation coverage is improved, but the reference lines become discontinuous at stitch points
Solution Approach 1:
By pre-identifying reference points with minimal curvature along path segments before high-speed traversal, the system prepares optimal connection points in advance. This preliminary selection ensures that when reference lines are stitched during high-speed operation, the connections are smooth and continuous, maintaining navigation stability without compromising speed
Solution Approach 2:
The patent applies local quality by selecting specific locations (points with smallest curvature) along the path segments for reference line connections. Rather than using uniform spacing, the system identifies local optimal points where curvature is minimal, ensuring that the local connection quality at stitch points maintains overall reference line smoothness even during high-speed navigation
Data Source
AI summary
In one embodiment, a method for generating a reference line for operating an autonomous driving vehicle includes determining a first ending reference point having a smallest curvature among a plurality of points within a first defined distance along a path, generating a first reference line based on a first initial reference point and the first ending reference point, determining a second ending reference point having a smallest curvature among a plurality of points within a second defined distance along the path, generating a second reference line based on the first and second ending reference points and an end section of the first reference line, connecting the first and second reference lines, and controlling the autonomous driving vehicle along the connected first reference line and the second reference line.